The present invention relates to a method of detecting pressure loss in vehicle tires.
The prior published application No. WO 95/12498 discloses the learning of correction factors with respect to the individual wheels of a vehicle during a first phase which shall compensate for possible differences in the diameters of the individual wheels. In the current operation the established wheel rotational speeds are multiplied with the respective learned correction factors in order to determine the vehicle speed from the wheel rotational speeds. The correction factors shall also be actually determined during driving. When the currently determined correction factor differs from the correction factor established for this wheel in the learning phase by more than a defined threshold value, pressure loss in the vehicle tire is detected. The correction factors are acquired in response to the driving situation so that the effects of possible deviations of wheel rotational speeds are taken into account which are due to longitudinal or transverse acceleration, due to cornering and the possibly related oversteering or understeering of the vehicle, or due to different slip values caused by differing coefficients of friction on the left and right road sides.
Further, it is known from U.S. Pat. Nos. 4,876,528 and 5,192,929 to detect pressure loss in tires of a vehicle by summing up the rotational speeds of the diagonal wheels and by establishing the difference between the sums. When this difference ranges between 0.05% and 0.6% of the mean value of both sums, the rotational speed of each individual wheel is allowed to deviate at most 0.1% from the mean speed of all four wheels. Pressure loss is signalled in the event of a greater deviation. Monitoring of the pressure loss is performed only if the vehicle is travelling straight, at least approximately. In the contrary case, the difference adopts a too great value due to a possibly understeering or oversteering behavior of the vehicle during cornering.
An object of the present invention is to provide a method which permits identifying pressure loss in a tire in a most simple and nevertheless maximum reliable fashion.
According to the present invention, this object is achieved because pressure loss in vehicle tires is identified by using at least one reference value which is produced by dividing the sums of respectively two signals representative of the wheel rotational speeds.
For example, the reference values can be produced as follows:
Ref1=(Tvl+Thr)/(Tvr+Thl)
Ref2=(Tvl+Tvr)/(Thr+Thl)
Ref3=(Tvl+Thl)/(Tvr+Thr).
The quantity Ti is the signal representative of the rotational speed of the wheel, and the indices i have the following meaning: vl=front left-hand, vr=front right-hand, hl=rear left-hand, hr=rear right-hand, and refer to the wheel concerned. In a particularly favorable manner, the quantity Ti as the signal representative of the rotational speed of the wheel can be established by the implementation of the method of the invention.
Advantageously, numerical problems in determining the reference values are minimized by the method of the present invention. In the prior art, initially two quantities are summed which have almost the same value. Thus, differences will occur only after a few digits behind the decimal point, from which, however, first problems may arise. In another step, the difference between these two sums is produced in the state of the art. As stated hereinabove, the two sums also adopt rather exactly the same value. In any case, deviations of these two sums inhere already a certain inaccuracy. When, on top of that, these two quantities are differenced, possible inaccuracies will still rise considerably so that the established difference, under certain circumstances, is no longer apt to be utilized as a numerical value.
According to the present invention, the reference values produce the sums of respectively two signals representative of the wheel rotational speeds, and respectively two of these sums are divided. Advantageously, this permits minimizing the numerical problems because the method of the present invention obviates the need for summing two almost equally great quantities and thereafter subtracting the sums in the subsequent steps.
In a preferred embodiment, the signals representative of the rotational speeds of the individual vehicle wheels are produced because the time measurement of the individual wheels, after a joint start signal, is started individually with the respectively following signal edge initiated by the respective wheel and is concluded individually for each wheel upon termination of a full or partial number of wheel revolutions predetermined for all wheels. In systems which make available only the number of the signal edges and the time of the last signal edge per scanning interval, favorably, an interpolation is made with reference to the corresponding signal edge which corresponds to the end of the measuring section.
Further inaccuracies can be avoided compared to the procedure of determining wheel rotational speeds which observes how many signal edges pass by in a predefined time window. In this mentioned procedure, there is an inaccuracy at the beginning and the end of the time window, in which position the sensor wheel is placed with respect to the last signal edge passing. This is because it may be the case that the last signal edge passage has just about taken place, or that the last signal edge passage took place so long ago that the following signal edge passage is imminent. Consequently, matters depend on the position which the sensor wheel adopts in the interspace between two teeth of the sensor wheel opposite the sensor.
The method of the present invention advantageously minimizes this inaccuracy because the position of the sensor wheel with respect to the sensor is exactly fixed at the beginning and the end of the measuring operation. With a sensor wheel having 48 teeth, this means in each case a {fraction (1/48)} rotation at the beginning and the end of the measuring operation.
The number of wheel rotations is predetermined as a function of the vehicle speed.
This provision may be effected, for example, so that the number of the wheel rotations varies proportionally with the vehicle speed. It may e.g. be achieved that the measurement time in which the signal representative of the wheel rotational speed shall be established is maintained approximately constant.
The number of the wheel turns can be integral. It is also possible to predefine a number of wheel turns which is not integral. In this case, the measuring operation ends with the edge of a tooth other than the tooth where measurement began. In order to begin the measuring operation exactly with one edge and to have it end exactly with the edge of another tooth, it is necessary to fix that part of the rotations which is no more completed in conformity with the number of the teeth of the sensor wheel.
Further embodiments of this procedure may include that for sensors with scanning intervals (which are so rated that the time information is not gathered for each signal edge but in which sensors scanning intervals are opened and closed with individual signal edges, and the number of the signal edges disposed in the scanning interval and the beginning and end time of the scanning interval is stored) an interpolation with reference to the defined signal edge is effected unless this edge marks the end of a scanning interval. The interpolation may take place under the assumption of a constant speed in the scanning interval, for example.
Measuring can be effected for each individual wheel. A general start signal is issued. After this start signal, the measuring operation is started individually for each wheel when the corresponding edge of the tooth of the sensor wheel has been detected. A defined edge of a tooth may e.g. be detected because a tooth gap is provided at one point of the sensor wheel in order to permit detecting the absolute angular position of the sensor wheel. Starting from this gap, the teeth identified can be included in the count and, thus, detected individually. However, it is also possible to arrange for other procedures of detecting and allocating the single teeth of the sensor wheel.
In a preferred embodiment, at least one comparative value can be predefined by adding an offset to the mean value of the reference value, or subtracting an offset therefrom, when a quality gauge of the scattering of the mean values does not exceed a defined value and/or when the established maximum and minimum values of the reference values differ by not more than a defined value from the mean value of the reference values.
Favorably, the comparative values are set when the established data show that only little variations in the reference values occur. The reference values in an unaccelerated straight travel are equal zero in the ideal case. Thus, a check can be made whether the reference values differ from the value 1 by not more than a defined threshold.
The quality gauge for scattering can be calculated from the filtered and unfiltered values and can be the standard deviation, for example.
Taking the filtered values into consideration individually permits identifying single or several measured values of the rotational speed with a major deviation which, with a comparatively great reliability, are not due to signal noise of the transducer because the noise effects have been filtered by the low-pass filter means. Favorably, the filter is a low-pass filter of first order. Detection of these major deviations eliminates the need to produce a comparative value in case excessive differences prevail with those values from which the comparative value shall be produced. Expediently, the production of a comparative value is suppressed when, caused by driving conditions, the rotational speeds of the wheels have different values as can be the case e.g. during cornering.
Taking into account the unfiltered values (favorably by way of integrating several values in a mean value and determining a quality gauge, e.g., the standard deviation) permits determining to what extent the wheel rotational speeds have changed in the interval during which the measured values were acquired.
In a preferred method, a first set of measured values and a second set of measured values is produced, and comparative values are produced as a function of the maximum value and the minimum value of the reference values when in both sets of measured values the maximum values of the reference values, the minimum values of the reference values and the respective arithmetical mean of the reference values of the two sets differ from each other by not more than respectively predetermined defined amounts and the quality gauge of scattering does not exceed a defined value.
Appropriately, this method is employed when excessively great variations occur continuously which render a setting of the comparative values impossible. The method of the present invention achieves an adaption of the comparative values to the extent of the variations of the reference values. More specifically, the variations of the reference values on an increased level are verified for stability. The comparative values are then determined as a function of the variations.
This permits adapting the accuracy requirements step by step to the manufacturing tolerances and deviations due to the driving conditions. The value of the offset may amount to 0.5 pro mille. The predetermined amount may be in the magnitude of 0.1 pro mille, and the other predetermined amount may also be in the magnitude of 0.1 pro mille. In a preferred aspect, the multiple of the quality gauge can be four times the quality gauge.
When a top and a bottom limit value is fixed in such a fashion, pressure losses are checked by means of these values. The top and bottom limit values are then maintained until these values are reset.
When the determination of the comparative values has not yet been successful in the second set of measured values, this method is repeated, and the corresponding values of the second set of measured values are taken into account for comparison in the next attempt.
In a preferred embodiment, the comparative values are determined axle-responsively.
This proves expedient especially when utilizing the reference value Ref2 alone or in combination. Where the objective is to detect pressure loss on both axles, modifications of the pressure loss bound limits can be permitted for the respective axle at issue. In the wheels of the driven axle, pressure losses cause a smaller deviation of the wheel rotational speed than in the wheels of the non-driven axle. Consequently, the bound limit, starting from which pressure loss is detected, is set to be lower in the wheels of the driven axle than in the wheels of the non-driven axle.
The comparative values are determined as a function of speed wherein, it has proven favorable to establish comparative values as supports, for example, in steps of 30 km/h to 50 km/h. In the routine operation, for example, the comparative value can be used whose speed is closest to the actual speed, or interpolation can be made between two comparative values of the next lower and next higher speed.
In a preferred embodiment, the values found are stored when driving is interrupted and read out again after restart of driving. In this arrangement, it is favorable that also a sneaking pressure loss, with the vehicle at standstill, can be detected immediately upon restart of driving. Further, the information about the air pressure in the vehicle tires is available instantaneously upon restart of driving. The memorized values are at least the comparative values.
A reset key is actuated when a wheel is changed because the air pressure in the tire during wheel change is most probably different compared to the wheel that was previously mounted. It is advantageous that the comparative values are redefined when a wheel change has been carried out.
If necessary, the redefining action can also be effected after check of the air pressure in the tire, which can be detected when the protective cap of the valve is removed.
A manually operable reset key is provided in a preferred embodiment. This also permits the vehicle driver to initiate a new determination of the comparative values when, for example, the air pressure of the tires was checked and correspondingly replenished. Reduction of the expenditure in terms of switches and sensors can be achieved thereby. It is then the duty of the user of the vehicle to ensure an initialization when a tire was changed or air was refilled.
Preferably, the production of the reference values is interrupted when defined threshold values of the longitudinal acceleration, transverse acceleration and/or wheel acceleration are exceeded. A test can be made by way of these quantities as to whether the wheel rotational speed signals show major deviations which are due to driving conditions such as cornering, where at least the rotational speeds of the curve-outward wheels differ from the rotational speeds of the curve-inward wheels. Further, when the vehicle exhibits an oversteering or understeering driving behavior in a bend, related deviations of the rotational speeds of the wheels of the front and rear axles will arise during cornering. It can be derived from the longitudinal acceleration whether deviations of the rotational speeds of the driven wheels from the rotational speeds of the non-driven wheels have to be expected which are due to an acceleration. It can be read from the wheel acceleration whether the vehicle moved through a pothole so that single wheels turn freely, with the result of a deviation of the rotational speed of these wheels. Interruption of the pick-up of measuring data is appropriate in all these cases.
Preferably, the production of the reference values is interrupted as soon as braking intervention is performed. This interrupts the acquisition of the values when the brake slip causes different rotational speeds of the wheels which are due to the braking operation. Braking intervention can be detected, for example, when an ABS system responds or when an active braking intervention is performed in a traction slip system. Likewise, this condition can be concluded from the brake light switch which responds already in braking operations where the amount of wheel slip is considerably smaller.
Thus, the method of the present invention favorably shows that one or a plurality of comparative values are determined only if it can be concluded from the established measured values that this is expedient. Compared to the other prior art method where the comparative value is newly established and, if necessary, rejected in a continuous way, the advantage achieved is that wheel rotational speeds which have already been falsified due to a changed tire pressure will no longer be taken into account in the comparative value. Because the rotational speeds of the wheels which have already suffered from pressure loss are also included in the production of the comparative value in the prior art method, the faulty wheel rotational speed value influences the comparative value whose special purpose is to detect that this wheel is failing. Thus, the comparative value is newly established in a continuous way in the prior art methods so that faulty wheel rotational speeds will automatically lead to a faulty comparative value. The detection of pressure loss is possible only with delay. In contrast thereto, the present method compares reference values with previously established comparative values so that pressure loss can be detected without delay.
An optical and/or audible warning signal may be issued as soon as pressure loss is detected.